Towards Simple and Accurate Human Pose Estimation with Stair Network

TL;DR

This paper introduces the STair Network, a small, efficient, and accurate human pose estimation model that outperforms larger models like HRNet while significantly reducing computational costs.

Contribution

The paper presents a novel small network architecture with feature diversity and local representations, plus two low-cost mechanisms for improved pose estimation accuracy.

Findings

Achieves 5.5% higher accuracy than HRNet on COCO with fewer parameters.

Reduces GFLOPs by 68% compared to HRNet.

Demonstrates effectiveness on standard pose estimation datasets.

Abstract

In this paper, we focus on tackling the precise keypoint coordinates regression task. Most existing approaches adopt complicated networks with a large number of parameters, leading to a heavy model with poor cost-effectiveness in practice. To overcome this limitation, we develop a small yet discrimicative model called STair Network, which can be simply stacked towards an accurate multi-stage pose estimation system. Specifically, to reduce computational cost, STair Network is composed of novel basic feature extraction blocks which focus on promoting feature diversity and obtaining rich local representations with fewer parameters, enabling a satisfactory balance on efficiency and performance. To further improve the performance, we introduce two mechanisms with negligible computational cost, focusing on feature fusion and replenish. We demonstrate the effectiveness of the STair Network on two standard datasets, e.g., 1-stage STair Network achieves a higher accuracy than HRNet by 5.5% on COCO test dataset with 80\% fewer parameters and 68% fewer GFLOPs.